Announcement

Collapse
No announcement yet.

General question for the sake of curiosity.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • General question for the sake of curiosity.

    Sabermetricians spend a lot of time struggling away with advanced math tools to come up with accurate ways to model the game of baseball...but your average fan of the game has little to know chance of understanding what good and bad values of those metrics are or putting any kind of magnitude on goodness or badness.

    If I tell you Babe Ruth Created 22 wins in 1921...your average baseball fan will say "22 wins...that sounds like a lot. Is that good?" It gets even muddier when I try to make the case that Edgar Martinez was worth more to his teams than say...Bill Mazeroski by claiming that Edgar produced 10-14 wins a year on offense alone on average, whereas Mazeroski would be lucky to create 6-8 wins combining his offense and defense. It's hard for someone to really place what "six wins created" means compared to 10...etc

    Fans of baseball think (unavoidably) in terms of easily identifyable statistics that are commonly published.

    I've put PCA win statistics into the form of batting averages...and am working on putting pitching statistics into the form of ERAs...but what kind of immediately recognizeable stat is there for representing fielding? I'm trying to figure out how to put PCA fielding wins into a form that is immediately recognizeable and that fans can just look at and know what I'm saying when I quote a statistic.

    What do you associate with fielding? I thought of F% but that is bounded on the high side be 1.000 and the distribution of F% is EXTREMELY screwed with a long, thick tail heading away from 1.000, making it unusable here. Do I have any other options?

  • #2
    So when the average fan sees raw OPS's of 1.13 and 1.03, they don't really understand the difference. But if you tell them that the first guy's OBP/SLG was .480/.650, and the second guy had .440/.590, they'd understand because those are common numbers we can judge from. So, is that your problem with the fielding % and why you won't use it? Cause you don't want to go over 1.000?

    Is there any way you can throw in a constant multiplier to get the final number under 1.000? You should look into that, cause for fielding, it's really the only stat out there for the average fan.

    Comment


    • #3
      --OTOH any serious fan knows FP tells you very little about defensive skill. And even people who think FP is a good indicator are unlikely to know offhand what a good one like they do BA or ERA. A 3B with a .975 is very surehanded, but a 1B with the same has hands of iron. There really are no defensive stats that strike a chord with the common fan. Actually, there are none that have universal appeal even to the uber fan.

      Comment


      • #4
        Originally posted by leecemark
        --OTOH any serious fan knows FP tells you very little about defensive skill. And even people who think FP is a good indicator are unlikely to know offhand what a good one like they do BA or ERA. A 3B with a .975 is very surehanded, but a 1B with the same has hands of iron. There really are no defensive stats that strike a chord with the common fan. Actually, there are none that have universal appeal even to the uber fan.
        Very true Mark.

        Most serious fans know that FP means very little, but the point is, that if you're a casual fan, you can look at a players FP alongside the lgFP and think you have a good idea of that players ability.

        I think what Matt is striving for, is that type of universal number; one which is on a scale that is immediately comprehended when glanced at in terms of what's "good" and "bad."

        Comment


        • #5
          Defense is really tough. Its virtually impossible to quantify with the constantly shifting situations of the game.

          Runners on, infield in, double play depth, shift, going to cover a base cause a runner breaks, bad hops, sun problems, differing outfield dimensions, official scorer discretion...

          and these are all things that are independent of the actual skill of the fielders.

          This mitigating circumstances are endless. You really have to watch the games and understand the situations as the present themselves. Most hardcore fans are well versed in the abilities of the players on the teams they root for but not as learned about the specific defnesive abilities of guys on other teams the way they are about the offensive abilities of the same players.

          I don't know if any actual defensive stat is held in high esteem by the common fan besides the error.
          THE REVOLUTION WILL NOT COME WITH A SCORECARD

          In the avy: AZ - Doe or Die

          Comment


          • #6
            Yeah...that's what I was afraid of.

            I have a few choices.

            I can use the ERA rubric, since defense and pitching combine to prevent runs. The problem with that is...fitting fielding to an ERA model might confuse people into thinking I'm claiming a team full of fielder Xs would have Y ERA, which would be very bad.

            I can use Range Factor. Most light-core fans at least know Range Factor exists...I would have to make it relative to position (eacfh position has a different RF distribution). But I don't know that a common fan can tell what a good and/or bad RF is.

            I can use batting average but invert it (good fielders get low batting averages)...

            I don't know...a solution doesn't immediately leap out at me.

            Comment


            • #7
              The best measure I think in terms of showing how good or bad a player is for say something like fielding would be to show the number compared to average on an over an under of 100. Like Park factors and others.

              For instance we could say that Bill Mazeroskis scores a 135 for secondbase fielding. What does that mean? Well Mazeroski for his career was 35% then the average second basemen.


              You will find that there are no standard fielding numbers but that can instantly identify as good or bad. There is nothing like batting average or ERA for fielding. If you say player A had a .958 fielding nobody is really going to know if that is good or bad. IF you say he a 3.68 range factor again nobody is really going to know.

              The other option would be to treat it like a video game and give out a rating of 0 to 100. Player A score a 98 for second base fielding player B and 88 so on and so on.

              Comment


              • #8
                Originally posted by SABR Matt
                I can use batting average but invert it (good fielders get low batting averages)...

                I don't know...a solution doesn't immediately leap out at me.
                Why can't you use BA, but not invert it. If the avg centerfielder in '39 had a defensive PCA of .300, and DiMaggio was at a .410, most would get that.

                I like the scale of 100 idea.

                Let's say the league average SS arm strength rating is 32/50, and the arm accuracy rating is 40/50. If we give Reyes a 40 and a 40, then his total would be 80/100 for the overall arm category, compared to a lg avg of 72/100. But then how do you account for the number of balls picked in the dirt that saved errors. And if he's less accurate when he goes to his right, or his left, or charges the ball.

                What about a right fielder who isn't that fast but makes up for it with other skills to be an above average fielder. We can't judge outfielder's arm strenth soley on assists, we use reputation and peers account for the most part. So how do you get that into fair numbers. Ok, just realized how impossible this is. Good luck.

                Comment


                • #9
                  Well defensive PCA doesn't set about rating individual elements that make up a fielder's job. It's a results oriented study. The methodology cares only about the plays made by a team of players, the runs that team allowed, the way in which those runs were allowed etc.

                  Rating defeense thogh is definitely the most complicated and delicate aspect of any sabermetric endeavor and the area where the most work still needs to be done.

                  I can certainly use a non-inverted batting average to explain defense...it just doesn't seem intuitive to me...but perhaps it's the best way to go...the non-SABR-oriented experts here can certainly tell me if they're understanding what is meant by PCA figures.

                  Comment


                  • #10
                    Very true, the average fan doesnt know about this stuff( I guess I'm above Average) but I think the new found age of fantasy baseball has made more baseball fans learn this stuff.

                    Comment


                    • #11
                      Flaw

                      Problem is, you are trying to take your metric and make it "casual fan friendly". Who cares? Either that fan is going to WANT to know and understand your stat or system (the way it is), or they will not want to have anything to do with it and just stick to triple crown categories.

                      Stick to your guns and your slide rule.
                      Your Second Base Coach
                      Garvey, Lopes, Russell, and Cey started 833 times and the Dodgers went 498-335, for a .598 winning percentage. That’s equal to a team going 97-65 over a season. On those occasions when at least one of them missed his start, the Dodgers were 306-267-1, which is a .534 clip. That works out to a team going 87-75. So having all four of them added 10 wins to the Dodgers per year.
                      http://www.youtube.com/watch?v=p5hCIvMule0

                      Comment


                      • #12
                        Even if fans could understand the scale- over 100 is good under is bad, or .400 is good and .200 is bad, they wouldn't use them because they don't know how they are calculated.

                        If you provide an explanation, general resopnse will be that it's too complex and that it doesn't make sense that players can provide "wins" because players aren't teams.

                        If it isn't easily understandable, it's wrong. And if it doesn't match ESPN's "expert analysis", it's wrong.
                        http://capitalfrontiers.com

                        Comment


                        • #13
                          Yeah redbuck...that kind of response has come up with casual fans a lot...I still have foolish hope that I can reach someone though if I make the effort...call me an optimist.

                          Comment


                          • #14
                            It has been my experience that the easiest of the non-basic stats to explain is anything based on 100 as average. I can manage to get through to my brother that if 100 is average, and 25% below that is bad, and 25% above that is good when talking about batting average or OBP, then I think that is the way to go for the "casual" fan. There will be some that ignore everything regardless....I call those the csh fans.

                            Comment


                            • #15
                              LOL...no...more like EH fans.

                              Comment

                              Ad Widget

                              Collapse
                              Working...
                              X